Image generation and editing
GPT Image 2.6 Flare
GPT Image 2.6 Flare is an OpenAI image generation and editing model for interactive workflows using text and image inputs.
Sizes 1024x1024, 1536x1024, 1024x1536, 2048x2048, 2048x1152, 1152x2048, 3840x2160, 2160x3840, auto · Quality auto, low, medium, high, xhigh, max · png · One image per request
$0.02076From per image
Price
- From per image
- $0.02076
- Per input image
- $0.00905
- Text input per 1M tokens
- $5.25
Includes the 5% MiniRouter fee. The final charge uses the provider’s actual cost and stays within your reserved hold. MiniRouter absorbs any overrun.
Estimate your credit hold
Credit is reserved before the request. The final charge may be lower.
Up to 32,000 characters.
Estimated hold
2× the total estimated cost, including the 5% fee. Your charge stays within this hold; unused credit is released.
Estimate only; nothing is sent. Includes text and one output image.
Use the Images API
Set MINIROUTER_KEY to your inference key. Select an operation and copy the example.
curl --fail-with-body --retry 0 --max-time 120 \
'https://api.minirouter.sh/v1/images/generations' \
-H "Authorization: Bearer $MINIROUTER_KEY" \
-D response.headers --output response.json \
-H 'Content-Type: application/json' \
--data '{
"model": "openai/gpt-image-2.6-flare",
"prompt": "A blue circle on a white background.",
"output_format": "png",
"quality": "low",
"size": "1024x1024",
"n": 1
}'Keep the request ID. The image response can arrive before its final charge. Check Request logs for the settled cost. A failed or disconnected request may still incur a cost; these examples disable automatic retries.
Supported operations
- Generate from a prompt
- POST /v1/images/generations
- Send JSON with a prompt. The response contains one base64 image.
- Edit an image
- POST /v1/images/edits
- Send one image up to 1024 × 1024 and a prompt as multipart form data.